Special Feature

A Limit to Bisphosphonate Treatment

Abstract & Commentary

By Leon Speroff, MD, Editor, Professor of Obstetrics and Gynecology, Oregon Health and Science University, Portland, is Editor for OB/GYN Clinical Alert.

Black and colleagues reported the results of an extension of the alendronate Fracture Intervention Trial (FIT) in which the participants were randomized after 5 years of treatment either to another 5 years of treatment or placebo.1 The group that discontinued alendronate treatment (5 or 10 mg per day) experienced average losses of bone mineral density over 5 years as follows: total hip - 2.4%, spine - 3.7%. The levels, however, remained above the pretreatment levels 10 years previously. There were no differences in nonvertebral fractures between the treatment and placebo groups. There was a twofold higher rate of clinically recognizable vertebral fractures in the placebo group (5.3% vs 2.4%).


The unique tight binding of bisphosphonates to bone matrix causes this drug to remain in the body for decades. This is believed to be the explanation for why there is no rapid bone loss after discontinuing bisphosphonate treatment in contrast to the rapid loss that follows the termination of estrogen therapy. This is also the reason why concerns have been raised regarding long-term treatment: because when bone remodeling releases bound bisphosphonate, it is free to be active again, and the result is that endogenous bisphosphonate is added to the administered bisphosphonate, raising dosage exposure. At this time, we don't know the lowest effective dose and the lowest effective duration of exposure. The potential risk that has been long recognized is that prolonged exposure to bisphosphonates or excessive dosage would oversuppress bone resorption, thus oversuppressing bone turnover and affecting the biomechanical strength of bone; indeed, allowing microcracks to accumulate.

Here is the critical point: Even though continued treatment produced a slightly greater bone density compared with the placebo group, the difference was small, and not meaningful clinically. The small difference indicates a residual effect of the previously administered alendronate that probably lasts longer than 5 years. The authors concluded that most women do not need long-term treatment, and that long-term treatment should be limited to high-risk women (women with existing vertebral fractures or very low bone densities).

It seems to me that several conclusions are warranted at this time:

  1. An increased susceptibility to nonspinal fractures may occur relatively early when bisphosphonate treatment is combined with another antiresorptive treatment (such as estrogen), and this should be avoided because no additional benefit or fracture risk has been demonstrated with combined treatment.
  2. Bisphosphonate treatment is best reserved for older postmenopausal women. It is not a drug of choice for the prevention of osteoporosis in relatively young postmenopausal women.
  3. In all except high-risk patients being treated with bisphosphonates, it would be wise to consider a 5-year time limit for duration of exposure, followed by monitoring of bone density, with a resumption of treatment in those who rapidly lose bone or in those who accumulate a loss of 5% to 10% in one year.


  1. Black DM, et al. JAMA. 2006;296:2927-2938.